Is Moore's law still true?

Why it makes a good exploration

Moore's law is quoted everywhere, usually without checking. Fitting an exponential model, extracting the doubling time and testing whether it has changed over different decades is a precise, answerable question.

The mathematics you'll need

• Exponential models and doubling time
• Logarithms to linearise data
• Linear regression on log-transformed data
• Comparing gradients across time periods

Course labels show where a technique sits; using maths from outside your course is fine if you explain it clearly and say it is new to you.

Where the data comes from

Download the “transistors per microprocessor” dataset from Our World in Data.

• Our World in Data — Clean country-level time series (CO₂, population, health, energy, education) with CSV download on every chart.

Cite every source in a footnote where you use it and in your bibliography. Check the licence of any dataset you download.

A possible outline

1. State Moore's claim precisely (doubling every two years).
2. Plot log₂ of transistor count against year.
3. Fit a line and convert the gradient to a doubling time.
4. Fit separate lines to each decade and compare doubling times.
5. Discuss what the log scale hides and what limits further growth.

Showing personal engagement

• Predict the transistor count of the processor in your own phone or laptop and look it up.
• Discuss whether a slowdown is visible and propose a model that captures it.
• Compare with another exponential claim in technology.

See Criterion C: personal engagement for what examiners look for.

Taking it further

Test whether a logistic or piecewise model fits significantly better, or compare with the growth of hard-drive capacity.